Project Overview
This notebook presents the project team’s exploratory data analysis (EDA) of credit card customers segmented by risk level. By combining stress scores, attention indicators, and customer segments, the team identifies priority groups for targeted marketing and risk management strategies.
Objectives:
- Quantify the proportion of customers needing attention
- Break down high-risk customers into meaningful segments
- Identify conversion opportunities for key at-risk groups
- Provide data-driven recommendations balancing growth and risk
Table of Contents
- Project Setup
- Library imports
- Data preparation
- Customer Risk Distribution
- Attention vs. no attention breakdown
- Segment Analysis
- Dormant / Casual Users
- Heavy Credit Users (Revolvers)
- Balanced Moderate Spenders
- Visualization
- Tree map representation
- Recommendations
| cust_id | balance | balance_frequency | purchases | oneoff_purchases | installments_purchases | cash_advance | purchases_frequency | oneoff_purchases_frequency | purchases_installments_frequency | cash_advance_frequency | cash_advance_trx | purchases_trx | credit_limit | payments | minimum_payments | prc_full_payment | tenure | cluster | customer_segment | utilization | min_pay_ratio | flag_cash_advance | flag_low_payment | flag_high_util | flag_min_only | flag_irregular_balance | stress_score | risk_level | need_attention | attention_status | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | C10001 | 40.900749 | 0.818182 | 95.40 | 0.00 | 95.4 | 0.000000 | 0.166667 | 0.000000 | 0.083333 | 0.000000 | 0 | 2 | 1000.0 | 201.802084 | 139.509787 | 0.000000 | 12 | 6 | Dormant / Casual Users | 0.040901 | 0.691320 | False | True | False | False | False | 1 | low | False | no attention needed |
| 1 | C10002 | 3202.467416 | 0.909091 | 0.00 | 0.00 | 0.0 | 6442.945483 | 0.000000 | 0.000000 | 0.000000 | 0.250000 | 4 | 0 | 7000.0 | 4103.032597 | 1072.340217 | 0.222222 | 12 | 0 | Heavy Credit Users (Revolvers) | 0.457495 | 0.261353 | False | True | False | False | False | 1 | low | False | no attention needed |
| 2 | C10003 | 2495.148862 | 1.000000 | 773.17 | 773.17 | 0.0 | 0.000000 | 1.000000 | 1.000000 | 0.000000 | 0.000000 | 0 | 12 | 7500.0 | 622.066742 | 627.284787 | 0.000000 | 12 | 2 | Balanced Moderate Spenders | 0.332687 | 1.008388 | False | True | False | True | False | 2 | low | False | no attention needed |
| 3 | C10004 | 1666.670542 | 0.636364 | 1499.00 | 1499.00 | 0.0 | 205.788017 | 0.083333 | 0.083333 | 0.000000 | 0.083333 | 1 | 1 | 7500.0 | 0.000000 | 312.343947 | 0.000000 | 12 | 6 | Dormant / Casual Users | 0.222223 | 0.000000 | False | True | False | False | False | 1 | low | False | no attention needed |
| 4 | C10005 | 817.714335 | 1.000000 | 16.00 | 16.00 | 0.0 | 0.000000 | 0.083333 | 0.083333 | 0.000000 | 0.000000 | 0 | 1 | 1200.0 | 678.334763 | 244.791237 | 0.000000 | 12 | 6 | Dormant / Casual Users | 0.681429 | 0.360871 | False | True | True | False | False | 2 | low | False | no attention needed |
| attention_status_count | |
|---|---|
| attention_status | |
| attention needed | 2412 |
| no attention needed | 6538 |
Our team calculated the distribution of customers requiring attention versus those who do not. The results show that approximately 26.9% of customers (2,412 out of 8,950) fall into the “needs attention” category. This proportion is significant and highlights a sizable group that warrants closer monitoring and targeted engagement strategies.
array(['attention needed'], dtype=object)
| customer_segment | attention_flag_count | attention_flag_percentages | |
|---|---|---|---|
| 0 | Dormant / Casual Users | 1051 | 44.0 |
| 1 | Heavy Credit Users (Revolvers) | 602 | 25.0 |
| 2 | Balanced Moderate Spenders | 471 | 20.0 |
| 3 | Low Balance, Low Engagement | 198 | 8.0 |
| 4 | Small High-Payment Anomalies | 54 | 2.0 |
| 5 | VIP High Spenders | 36 | 1.0 |
Within the 2,412 customers identified as needing attention, the majority fall into three key segments:
Dormant / Casual Users (~44%)
- Behavior: Low spend and engagement, minimal contribution to revenue.
- Team Conversion Goal: Increase transaction activity and engagement without increasing credit risk.
- Opportunities: Re-engagement campaigns, category-specific offers, gamification (spend milestones), partnerships with merchants, digital nudges.
- Risk Control: Maintain low credit exposure and focus on frequency rather than balance growth.
Heavy Credit Users / Revolvers (~25%)
- Behavior: High balances and cash advance usage, minimal repayments — profitable but high risk.
- Team Conversion Goal: Retain revenue while encouraging healthier repayment habits.
- Opportunities: Debt consolidation, flexible repayment tools, credit health programs, incentives to shift spend from cash advances to retail, financial coaching.
- Risk Control: Monitor repayment improvements and restrict further credit increases until stability is observed.
Balanced Moderate Spenders (~20%)
- Behavior: Moderate, predictable spending with low risk but untapped potential.
- Team Conversion Goal: Move toward premium or high-value segments (VIP spenders or full-payment customers).
- Opportunities: Tiered reward upgrades, installment plans, lifestyle campaigns (travel, dining), cross-selling premium products, anniversary promotions.
📌 These findings provide the foundation for prioritizing customer conversion strategies that balance revenue growth with risk management.